NOTE: this page is for archival only, see the note at the end of the page.

mainline issues

mac80211

  • hide (or get rid of) master dev, use phy name for cfg80211
  • compat: handle reassociation when the sta is out of range for a moment and then comes back

AP support

AP support requires mac80211 and nl80211 enhancements. In addition to this you need external wireless-test.git patches from johill and hostapd patches. The following are items we need to handle to make AP support IEEE-802.11 specification compliant:

  • short slot
  • short preamble
  • basic rate set
  • rts/cts
  • channel flag/regulatory domain
  • tx queue params (the plan is to make them qdisc settings after QoS patch)
  • multi-bss support (no kernel changes, just hostapd keep track of BSSes)
  • internal bridge setting (separated STAs in AP, local->bridge_packets)

  • CTS-protect setting for AP
  • check why we see data frames on cooked monitor
  • investigate the skb truesize bug

Intel is working (and already completed in their own tree) these features:

  • 802.11n
  • WMM AC
  • u-APSD support

cfg80211

  • Provide a replacement for all iwconfig functionality
  • /sys/class/net/$foo/wireless is managed by Wireless-Extensions (in net/core/net-sysfs.c). This directory has this information:

mcgrof@monster:~/devel/compat-wireless-2.6$ ls -l /sys/class/net/wlan0/wireless/
total 0
-r--r--r-- 1 root root 4096 2007-12-10 16:53 beacon
-r--r--r-- 1 root root 4096 2007-12-10 16:53 crypt
-r--r--r-- 1 root root 4096 2007-12-10 16:53 fragment
-r--r--r-- 1 root root 4096 2007-12-10 16:53 level
-r--r--r-- 1 root root 4096 2007-12-10 16:53 link
-r--r--r-- 1 root root 4096 2007-12-10 16:53 misc
-r--r--r-- 1 root root 4096 2007-12-10 16:53 noise
-r--r--r-- 1 root root 4096 2007-12-10 16:53 nwid
-r--r--r-- 1 root root 4096 2007-12-10 16:53 retries
-r--r--r-- 1 root root 4096 2007-12-10 16:53 status

All of the information on each file is obtained through a call to dev->wireless_handlers->get_wireless_stats(dev). We don't want to break old drivers but we do want to stop using WE as soon as cfg80211 provides sufficient API for its replacement. When that happens drivers can use its own cfg80211 call instead of get_wireless_stats.

We should probably not add more information to this directory and instead look towards configfs.

mac80211

Most important issues

  • Need to stop TX/RX when a radar is detected for the duration of scan for a new channel. [reported by Jouni Malinen]

Other issues

  • Add a new notification type to the ieee80211_notify_mac function that drivers can call when they resume. It should

    • call set_key for each key that was uploaded to hardware
    • reconfigure hardware (channel etc)
    • reconfigure interfaces
    • ...?

    (when done please remove from http://kernelnewbies.org/KernelProjects too)

scanning

  • possibly make scanning independent of virtual interface (only use MAC address of the interface that the scan was triggered for) – JohannesBerg 2007-09-24 12:53:39

core

misc

  • Export more hardware capabilities as wiphy attribute via nl80211.
  • Easier handling of configuration requests (ieee80211_hw->config()). In the current implementation, driver has no way to find out which parameter is being set; setting all parameters on every call to config() is obviously not a good idea.

  • Kicking DMA on the last fragment only (?) [suggested by Ivo van Doorn]
  • Why there is a beacon_update() handler? Maybe it would be easier for the driver to be allowed to call ieee80211_beacon_get() even in IBSS mode.
  • Move ETH_P_PAE from ieee80211_i.h to linux/if_ether.h.
  • Make IEEE80211_FRAGMENT_MAX configurable (preferably at run-time).
  • Fix the race in PS status update (see TODO in ieee80211_tx_status()).
  • Sort out function definitions in ieee80211_i.h (they are at two different places in the file now).
  • ieee80211_if_conf should be a part of ieee80211_sub_if_data
  • In case of STA or IBSS, after a change of SSID or generic_element, ieee80211_if_conf should be called.
  • Try to switch from sw crypto to hw when there was no more room for STA keys in the hardware and one STA disassociates (so there is possibly a room now). [suggested by Johannes and Michael]

qdisc bugs

  • can't put the ieee80211 qdisc back on while the device is running

From Patrick McHardy:

  • classify_1d doesn't care about tc_classify return values. tc_classify may decide to steal packets, drop them, etc. In case of stolen packets this causes use-after-free, otherwise just malfunctions.
  • classify_1d returns res.class if it is != -1, which can never happen (except with an empty classifier list because of the explicit initialization, but you should check the return code) since ->get() and ->bind_tcf() both return 0 for invalid classes and the classid otherwise. There's also an off-by-one, classids start at one, so it should return res.class - 1 (or better res.classid - 1, which is meant to be a numerical identifier).

Considering that it is possibly and may be desirable to attach a different qdisc than the built-in multiband qdisc, it might also make sense to split the 80211 specific classification in a seperate classifier module to allow simple classification of management traffic with other qdiscs.

Library functions we should add

  • PLCP Length calculation
  • Functions to fill modes/rates for B, BG, and ABG.

Optimizations

  • ieee80211_get_hdrlen and ieee80211_is_eapol are called very often.
  • Recognition of device incoming frame belongs to can be made much smarter and faster.
  • When one packet is dumped through several interfaces, some operations can be performed just once (e.g. searching for the key, sometimes decrypting, defragmentation etc.).
  • think about handling probe responses in firmware like b43 can. This requires telling hostapd that it shouldn't be replying to probe requests and having it give the appropriate info to the kernel [Johannes/Michael]
  • Order available BSSes by rssi in STA/IBSS mode. We do this in ieee80211_rx_bss_add() ieee80211_sta.c

possibly in the future

  • Add #ifdef's for not compiling AP stuff. [suggested by Jouni Malinen]

Coding style

  • remove forward declarations

userspace mlme

Goal

The end goal is to have a communication path as follows:

Legend:

nm:        Network Manager               - GUI based utility
iw:        Wireless-tool replacement     - Console based utility
           (these are prototypical, there will be more tools)
nl80211:   netlink-based wire-format for wireless configuration
umlme:     Userspace MLME (wpa_supplicant)
fullmac:   non-mac80211 full-mac hardware driver
cfg80211:  kernel wireless config agent speaking to drivers/mac80211 and userspace
---

Communication with userspace

  Hardware and kernel/hardware MLME configuration:
                           nl80211
    { nm | iw | umlme } -------------> { cfg80211 }


  Userspace MLME configuration (alternative 1)
                   nl80211                     nl80211
    { nm | iw } -------------> { cfg80211 } -------------> { umlme }

  Userspace MLME configuration (alternative 2)
                   nl80211
    { nm | iw } -------------> { umlme }

  Userspace MLME configuration (alternative 3)
                  socket (*)
    { nm | iw } -------------> { umlme }

    (*) Unix socket is just one alternative, could be a FIFO too
        or a TCP socket or a d-bus based protocol or ...

Communication inside the kernel

                 cfg80211_ops
    { cfg80211 } -------------> { fullmac | mac80211 }

Current status

wpa_supplicant currently uses many ioctls and the management interface both of which will not go into mainline in this form.


This is a static dump of the wiki, taken after locking it in January 2015. The new wiki is at https://wireless.wiki.kernel.org/.
versions of this page: last, v359, v358, v357, v356, v355, v354, v353, v352, v351, v350, v349, v348, v347, v346, v345, v344, v343, v342, v341, v340, v339, v338, v337, v336, v335, v334, v333, v332, v331, v330, v329, v328, v327, v326, v325, v324, v323, v322, v321, v320, v319, v318, v317, v316, v315, v314, v313, v312, v311, v310, v309, v308, v307, v306, v305, v304, v303, v302, v301, v300, v299, v298, v297, v296, v295, v294, v293, v292, v291, v290, v289, v288, v287, v286, v285, v284, v283, v282, v281, v280, v279, v278, v277, v276, v275, v274, v273, v272, v271, v270, v269, v268, v267, v266, v265, v264, v263, v262, v261, v260, v259, v258, v257, v256, v255, v254, v253, v252, v251, v250, v249, v248, v247, v246, v245, v244, v243, v242, v241, v240, v239, v238, v237, v236, v235, v234, v233, v232, v231, v230, v229, v228, v227, v226, v225, v224, v223, v222, v221, v220, v219, v218, v217, v216, v215, v214, v213, v212, v211, v210, v209, v208, v207, v206, v205, v204, v203, v202, v201, v200, v199, v198, v197, v196, v195, v194, v193, v192, v191, v190, v189, v188, v187, v186, v185, v184, v183, v182, v181, v180, v179, v178, v177, v176, v175, v174, v173, v172, v171, v170, v169, v168, v167, v166, v165, v164, v163, v162, v161, v160, v159, v158, v157, v156, v155, v154, v153, v152, v151, v150, v149, v148, v147, v146, v145, v144, v143, v142, v141, v140, v139, v138, v137, v136, v135, v134, v133, v132, v131, v130, v129, v128, v127, v126, v125, v124, v123, v122, v121, v120, v119, v118, v117, v116, v115, v114, v113, v112, v111, v110, v109, v108, v107, v106, v105, v104, v103, v102, v101, v100, v99, v98, v97, v96, v95, v94, v93, v92, v91, v90, v89, v88, v87, v86, v85, v84, v83, v82, v81, v80, v79, v78, v77, v76, v75, v74, v73, v72, v71, v70, v69, v68, v67, v66, v65, v64, v63, v62, v61, v60, v59, v58, v57, v56, v55, v54, v53, v52, v51, v50, v49, v48, v47, v46, v45, v44, v43, v42, v41, v40, v39, v38, v37, v36, v35, v34, v33, v32, v31, v30, v29, v28, v27, v26, v25, v24, v23, v22, v21, v20, v19, v18, v17, v16, v15, v14, v13, v12, v11, v10, v9, v8, v7, v6, v5, v4, v3, v2, v1